394 Commits (8da75fb272f1724c4862c54fc61686c1a4793eb4)

Author SHA1 Message Date
Michael Niedermayer 4e1349d4e1 use fastmemcpy for chrominance if no chrominance filtering is done 23 years ago
Michael Niedermayer cd38e322ef faster dering 23 years ago
Michael Niedermayer 043ba56f68 1% speedup 23 years ago
Michael Niedermayer cf5ec61d0c oops, too much cleanup ;) 23 years ago
Michael Niedermayer c09dc46524 cleanup 23 years ago
Michael Niedermayer 7f16f6e64e faster mmx2 / 3dnow deblocking filter 23 years ago
Felix Bünemann 652cb0f57a Move ignores in subdirs to their respective .cvsignore files. 23 years ago
Gabucino 21810797bc postproc/ dir now respects --with-extraincdir 23 years ago
Michael Niedermayer a9c779788f temp denoiser: 23 years ago
Michael Niedermayer be44a4d74a temp denoiser changes: (a-b)^2 instead of |a-b| and MMX2/3DNOW version 23 years ago
Michael Niedermayer 117e45b03b temporal noise reducer in C (-pp 0x100000) 23 years ago
Michael Niedermayer ed8c06708e uyvytoyv12 in MMX (untested) 23 years ago
Michael Niedermayer 815cbfe71e QP_store==null bugfix and no opendivx bugfix 23 years ago
Michael Niedermayer ed3460658f yuy2toyv12 bugfix 23 years ago
Arpi 81c0590e6c uyvy->uv12 added 23 years ago
Michael Niedermayer cff6ecd7e0 .balign or we´ll align by 64kb on some architectures 23 years ago
Michael Niedermayer 6e3bba7282 .align 16 23 years ago
Pierre Lombard 6c426cff79 added missing #include <malloc.h> 23 years ago
Michael Niedermayer 9395185f9e bugfix 23 years ago
Nick Kurshev 1b383194f6 rgb32tobgr32 now is documented ;) 23 years ago
Michael Niedermayer 9996924345 rgb32tobgr32 / bgr32torgb32 23 years ago
Michael Niedermayer 3fe8e8f068 gcc does optimize writes to non volatile variables away if it didnt know that they were read in between 23 years ago
Michael Niedermayer da3b92c877 15/16 bpp dithering 23 years ago
Michael Niedermayer d8fa3c5497 better dithering 23 years ago
Nick Kurshev 2b3eef22d0 Minor speedup of rgb32to24. (performance is not successful) 23 years ago
Jürgen Keil 9e7f791dcb Remove a few obsolete header includes (fixes a compile error on solaris) 23 years ago
Michael Niedermayer b28daef82e minor changes (fixed some warnings, added attribute aligned(8) stuff) 23 years ago
Nick Kurshev 53445e8301 MMX2, 3DNOW, MMX optimized rgb32(24)to16(15) stuff 23 years ago
Nick Kurshev 3eb2151c15 optimized rgb24to16 stuff 23 years ago
Michael Niedermayer 761635727d mmx2 bgr24 stuff from swscale (slightly faster) 23 years ago
Nick Kurshev 0155db7f6c rgb24to32 now is faster 23 years ago
Michael Niedermayer 18e19c3159 vars are aligned by 8 now (even faster, yes they where not aligned correctly) 23 years ago
Michael Niedermayer 7ad16638de 16bpp is 10% faster 23 years ago
Michael Niedermayer 607379b38b 15 bpp support 23 years ago
Michael Niedermayer 574406bb06 common code moved to a #define 23 years ago
Nick Kurshev 6611aa83de Move yuv2rgb to postprocess 23 years ago
Michael Niedermayer 99d2cb7219 faster bgr24 output 23 years ago
Michael Niedermayer bdc2eb9a00 only 6 registers used 23 years ago
Michael Niedermayer cfc15dc6bc unsigned stuff 23 years ago
Michael Niedermayer dabcdbc493 yuy2toyv12 fixed and speedup 23 years ago
Michael Niedermayer 42b5fcb890 fixed yv12toyuy2 23 years ago
Nick Kurshev 4aecdf910f Forgot 23 years ago
Nick Kurshev 8dd6739c94 Fixed rgb32(24)to15 stuff 23 years ago
Nick Kurshev 90226a43e6 Fixed rgb32(24)to16 stuff, rgb32(24)to15 is still broken 23 years ago
Nick Kurshev 996e1a7cff Well - old algorithms and new stuff rgb24to16(15) 23 years ago
Nick Kurshev 12e3287f03 More standards compilance 23 years ago
Michael Niedermayer bd09433fa9 yuy2toyv12 in MMX 23 years ago
Michael Niedermayer 4060205ba7 yv12toyuy2 in MMX 23 years ago
Michael Niedermayer d9d58d172e yv12 <-> yuy2 in C 23 years ago
Michael Niedermayer 9b2c28e6ed faster 15to16 bit rgb (the mmx routine is limited by memory speed so there is no difference ): but the c routine is faster 23 years ago